Bug 701879

Summary: auto install with cciss bus_type fail at disk partition
Product: Red Hat Enterprise Linux 6 Reporter: Mohua Li <moli>
Component: ovirt-nodeAssignee: Mike Burns <mburns>
Status: CLOSED ERRATA QA Contact: Virtualization Bugs <virt-bugs>
Severity: high Docs Contact:
Priority: high    
Version: 6.1CC: apevec, cshao, gouyang, leiwang, mburns, ovirt-maint, syeghiay, ycui, zdover
Target Milestone: rc   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: ovirt-node-2.0.2-0.2.git1b3c2f5.el6 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2011-12-06 19:14:58 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
ovirt.log
none
cciss interactive failed at partition
none
Patch mburns: review? (apevec)

Description Mohua Li 2011-05-04 07:24:51 UTC 
Description of problem:

boot with pxe specify storage_init=cciss,storage_vol=::::: local_boot firstboot, fail at creating root & rootbackup,


mke2fs 1.41.12 (17-May-2010)

/dev/cciss/c0d0p1 is apparently in use by the system; will not make a filesystem here!

+ __st=1

+ stop_log

+ '[' 0 '!=' 1 ']'

+ exec

+ exec


[root@hp-slot1-eth0 log]# find /sys -name *50:08.0*

/sys/devices/pci0000:40/0000:40:11.0/0000:4f:00.0/0000:50:08.0

/sys/bus/pci/devices/0000:50:08.0

/sys/bus/pci/drivers/cciss/0000:50:08.0

++ sfdisk -s /dev/cciss/c0d0

+ local disk_kb_count=71652960

+ dd if=/dev/zero of=/dev/cciss/c0d0 bs=1024 seek=71652959 count=1

1+0 records in

1+0 records out

1024 bytes (1.0 kB) copied, 5.3467e-05 s, 19.2 MB/s

+ kpartx -d /dev/cciss/c0d0

+ '[' -L /dev/mapper/HostVG-Swap ']'

++ find '/dev/mapper/HostVG*'

find: `/dev/mapper/HostVG*': No such file or directory

+ reread_partitions /dev/cciss/c0d0

+ local drive=/dev/cciss/c0d0

+ udevadm settle

+ [[ /dev/cciss/c0d0 =~ /dev/mapper ]]

+ set +e

+ blockdev --rereadpt /dev/cciss/c0d0

+ set -e

+ log 'Creating Root and RootBackup Partitions'

++ date '+%b %d %H:%M:%S'

+ printf 'May 04 06:19:37 '

May 04 06:19:37 + '[' 0 = 1 ']'

+ echo 'Creating Root and RootBackup Partitions'

+ let 'RootBackup_end=256*2'

+ true

+ set +e

+ parted -s /dev/cciss/c0d0 'mklabel msdos mkpart primary ext2 2048s 256M mkpart primary ext2 256M 512M'

+ set -e

+ reread_partitions /dev/cciss/c0d0

+ local drive=/dev/cciss/c0d0

+ udevadm settle

+ [[ /dev/cciss/c0d0 =~ /dev/mapper ]]

+ set +e

+ blockdev --rereadpt /dev/cciss/c0d0

+ set -e

+ '[' -e /dev/cciss/c0d02 -o /dev/cciss/c0d0p2 ']'

+ break

+ partroot=/dev/cciss/c0d01

+ partrootbackup=/dev/cciss/c0d02

+ '[' '!' -e /dev/cciss/c0d01 ']'

+ partroot=/dev/cciss/c0d0p1

+ partrootbackup=/dev/cciss/c0d0p2

+ ln -snf /dev/cciss/c0d0p1 /dev/disk/by-label/Root

+ mke2fs -t ext2 /dev/cciss/c0d0p1 -L Root

mke2fs 1.41.12 (17-May-2010)

/dev/cciss/c0d0p1 is apparently in use by the system; will not make a filesystem here!

+ __st=1

+ stop_log

+ '[' 0 '!=' 1 ']'

+ exec

+ exec

steps,




Version-Release number of selected component (if applicable):
rhev-hypervisor 6.1-20110427

How reproducible:
always with cciss bus_type 

Steps to Reproduce:
1,boot with pxe specify storage_init=cciss,storage_vol=::::: local_boot firstboot

2.
3.
  
Actual results:


Expected results:


Additional info:

pls see bug 691891 comment 26,

tested with storage_init=(usb,scsi,ata,cciss) all could be parse successfully,
and installed successfully expect cciss, which has the error refer in comment
21, will open a new bug for it,
Comment 1 Mohua Li 2011-05-04 07:26:53 UTC 
Created attachment 496715 [details]
ovirt.log
Comment 11 Mike Burns 2011-08-17 16:22:13 UTC 
It appears that this is happening because the cciss device is picked up by multipath, but is not handled correctly when translating to /dev/mapper.
Comment 12 Mike Burns 2011-08-17 17:16:15 UTC 
Created attachment 518722 [details]
Patch

Problem was related to translating /dev/cciss/c0d0 device to its multipath equivalent.  The install failed because multipath was holding the /dev/cciss/c0d0 device busy.

For testing:

autoinstall to clean cciss device
autoinstall to cciss device with previous install of rhevh
Comment 16 Zac Dover 2011-10-12 06:41:06 UTC 
    Technical note added. If any revisions are required, please edit the "Technical Notes" field
    accordingly. All revisions will be proofread by the Engineering Content Services team.
    
    New Contents:
A previous version of Red Hat Enterprise Virtualization Hypervisor did not properly install when installing through pxe and specifying CCISS (a block driver for older HP Smart Array RAID controllers), because the cciss device was not handled correctly when translating to /dev/mapper. Multipath held the /dev/cciss/c0d0 device busy.

This issue has been fixed, and you can now install with CCISS.
Comment 17 Alan Pevec 2011-10-25 12:33:33 UTC 
Deleted Technical Notes Contents.

Old Contents:
A previous version of Red Hat Enterprise Virtualization Hypervisor did not properly install when installing through pxe and specifying CCISS (a block driver for older HP Smart Array RAID controllers), because the cciss device was not handled correctly when translating to /dev/mapper. Multipath held the /dev/cciss/c0d0 device busy.

This issue has been fixed, and you can now install with CCISS.
Comment 18 errata-xmlrpc 2011-12-06 19:14:58 UTC 
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

http://rhn.redhat.com/errata/RHBA-2011-1783.html